© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 30012
    Prev
    Next
    INDEX
    Explanations

    help me or with

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    uers
    -0.09
    teenth
    -0.09
    å¿Į
    -0.08
    Floor
    -0.08
    á»ĵng
    -0.08
    Pale
    -0.08
    ogle
    -0.08
    erate
    -0.08
    vice
    -0.08
    åľī
    -0.08
    POSITIVE LOGITS
     Äijỡ
    0.20
    fully
    0.18
    desk
    0.17
     ÃŃch
    0.17
    lessness
    0.15
    lessly
    0.15
    ðŁı»
    0.13
    à¹Ģหล
    0.13
    fulness
    0.12
    ful
    0.10
    Activations Density 0.090%

    No Known Activations